Neural based optimization analysis of distributed MEMS transmission line phase shifters

Abstract

RF MEMS (Radio Frequency Micro Electro Mechanical System) technology is a key innovation for building low-loss phase shifters and other control circuits at millimeter-wave frequencies. The developments of electronic phase shifters with high insertion loss have been used in phased array radar and in wide range of systems including communications and measurement instrumentation. In this paper we propose an efficient approach based on Artificial Neural Network (ANN) for optimization of Distributed MEMS Transmission Line (DMTL) to obtain the maximum amount of phase shift with minimum insertion loss. Based on the work of Rodwell et al for analysis of loss in distributed non linear Coplanar Waveguide (CPW) line, we extend the work for optimization for best phase shift. A stand -alone model for optimization of DMTL phase shift is derived efficiently using ANN. The results from the neural models trained by Levenberg - Marquardt algorithm are in very good agreement with the theoretical results available in the literature.